Cilostazol prevents amyloid β peptide25-35-induced memory impairment and oxidative stress in mice

RESEARCH PAPERbph_1014 1899..1912

Cilostazol prevents amyloidb peptide25-35-inducedmemory impairment andoxidative stress in miceMasayuki Hiramatsu1,2, Osanao Takiguchi1, Aki Nishiyama2 and

Hiromasa Mori1,3

1Laboratory of Neuropsychopharmacology, Graduate School of Environmental and Human

Sciences, Meijo University, Nagoya, Japan, 2Department of Chemical Pharmacology, Faculty of

Pharmaceutical Sciences, Meijo University, Nagoya, Japan, and 3Otsuka Pharmaceutical Co., Ltd,

Tokyo, Japan

CorrespondenceMasayuki Hiramatsu, Laboratoryof Neuropsychopharmacology,Graduate School ofEnvironmental and HumanSciences, Meijo University,Tenpaku-ku, Nagoya 468-8503,Japan. E-mail:mhiramt@meijo-u.ac.jp----------------------------------------------------------------

Keywordscilostazol; amyloid b peptide25-35;oxidative stress; lipidperoxidation; malondialdehyde;memory----------------------------------------------------------------

Received25 January 2010Revised6 August 2010Accepted17 August 2010

BACKGROUND AND PURPOSECilostazol may be effective in dementia associated with a cerebral ischaemia. In this study, we examined whether it exertsbeneficial effects on learning and/or memory impairment induced by Ab25-35 in mice, and compared its effects with those ofaspirin.

EXPERIMENTAL APPROACHAb25-35 (9 nmol) was administered to mice i.c.v. Learning and memory behaviour were evaluated by measuring spontaneousalternation in a Y-maze and a step-down type passive avoidance test, on the 5th and 8th days after injection respectively.Levels of lipid peroxidation (malondialdehyde) and cytokines in the frontal cortex and hippocampus were measured 2, 3, 5and 7 days after the Ab25-35 injection. The effects of repeated administration of cilostazol and aspirin (both at 30 and100 mg·kg-1, p.o.) on any changes induced by Ab25-35 were evaluated.

KEY RESULTSRepeated administration of cilostazol significantly attenuated the impairment of spontaneous alternation and the shortenedstep-down latency induced by Ab25-35. Aspirin did not show any beneficial effect. A significant increase in the levels ofmalondialdehyde (MDA) and IL-1b (only measured in hippocampus) was observed 2, 3 and 5 days after the Ab25-35 injectionin the frontal cortex and hippocampus. Repeated administration of cilostazol (100 mg·kg-1) completely prevented the increasein MDA levels but failed to antagonize the increase in the expression of IL-1b induced by Ab25-35.

CONCLUSIONS AND IMPLICATIONSThese results suggest that the protective effect of cilostazol on Ab25-35-induced memory impairment may be related tooxidative stress in the frontal cortex and the hippocampus.

AbbreviationsAb, amyloid b; AD, Alzheimer’s disease; APP, amyloid precursor protein; BACE, b-site of amyloid precursor proteincleaving enzyme; CMC, carboxymethyl cellulose sodium salt; CREB, cAMP-responsive element binding proteincilostazol:6-[4-(1-cyclohexyl-1H-tetrazol-5-yl)butoxy]-3,4- dihydro-2(1H)-quinolinone; GM-CSF, granulocyte macrophagecolony-stimulating factor; IFN-g, interferon-g; IGF-1, insulin-like growth factor-1; IL, interleukin; IkB, inhibitor ofnuclear factor-kB; LPS, lipopolysaccharide; LTP, long-term potentiation; MCA, middle cerebral artery; MDA,malondialdehyde; NF-kB, nuclear factor kB; NO, nitric oxide; NSAIDs, non-steroidal anti-inflammatory drugs; PDE III,phosphodiesterase III; ROS, reactive oxygen species; TBARS, thiobarbituric acid-reactive substance; TNF-a, tumournecrosis factor-a; i.h., intrahippocampal; iNOS, inducible nitric oxide synthase

BJP British Journal ofPharmacology

DOI:10.1111/j.1476-5381.2010.01014.xwww.brjpharmacol.org

British Journal of Pharmacology (2010) 161 1899–1912 1899© 2010 The AuthorsBritish Journal of Pharmacology © 2010 The British Pharmacological Society

Introduction

It is well known that oxidative stress is involved inseveral diseases, such as neurodegenerative disordersand ischaemic encephalopathy (Pappolla et al.,1998; Choi et al., 2002). Alzheimer’s disease (AD) isa progressive neurodegenerative disorder associatedwith global mental dysfunction and impairmentof cognitive function (Palmer, 2002). Commonpathological features of AD include senile plaques,neurofibrillary tangles, and neuronal loss in regionsof the brain involved in learning and memoryfunctions (Rowan et al., 2007). In Alzheimer’s typedementia, amyloid b (Ab) protein such as Ab1-42,generates reactive oxygen species (ROS) and nitricoxide, which, by inducing hyperoxidation ofneuronal proteins and lipids, may be associatedwith learning and memory impairment (Fangand Liu, 2006). The active fragment of Ab protein,Ab25-35, has similar effects; therefore, learning andmemory impairment induced by Ab25-35 is alsomediated by ROS (Lu et al., 2009). In animal experi-ments, intrahippocampal or i.c.v. injection ofAb25-35 has been shown to induce histological andbiochemical changes; learning deficits (Mauriceet al., 1996; Meunier et al., 2006; Alkam et al., 2008);dysfunction of the cholinergic system, whichplays an important role in the cognitive deficitsassociated with aging and neurodegenerative dis-eases (Tran et al., 2001); and oxidative stress-mediated changes in hippocampal long-termpotentiation (Trubetskaya et al., 2003). Thus, Ab25-35-injected animals are one of the models often used toinvestigate the pathogenesis and progression of AD,and for evaluating new therapeutic agents for AD(Maurice et al., 1996).

Cilostazol is an antagonist of PDE III and is usedclinically as an antiplatelet drug. As cilostazolincreases the cerebral blood flow (Kwon et al., 2005)and decreases the size of cerebral infarcted area morestrongly than low doses of aspirin (Lee et al., 2005),it may be effective in treating the type of dementiaassociated with a decrease and stoppage of the cere-bral blood flow in brain blood vessels. Interestingly,cilostazol has been shown to inhibit lipid peroxida-tion and apoptosis in a model of cerebral ischaemia(Watanabe et al., 2006). Lipid peroxidation andapoptosis may be associated with the neuronaldysfunction caused by Ab25-35, including the alteredtranscription factors and the inhibition of inflam-matory cytokine release from macrophages (Kimet al., 2002). Furthermore, because cilostazol cantrap ROS and inhibit the cell damage caused byoxidative stress (Choi et al., 2002), it has an antioxi-dative effect and may be useful as a treatment for thedementia associated with AD. In addition, cilostazol

has been shown to be more beneficial in the main-tenance and improvement of memory function inAlzheimer’s patients currently taking the medica-tion donepezil (Arai and Takahashi, 2009). However,the ameliorative effect of cilostazol on the learningand memory impairment induced by Ab has notbeen studied, and its mechanism is not yet clear.

Recent reports have indicated that patients suf-fering from inflammatory diseases (e.g. arthritis)who take anti-inflammatory medication have areduced risk of developing AD (Montine et al., 1999;Sugaya et al., 2000). Clinical cohort studies havereported that the use of non-steroidal anti-inflammatory drugs reduces the risk of developingAD (Szekely et al., 2008). Similar to cilostazol,aspirin, because of its anti-platelet action via inhi-bition of the cyclooxygenase enzyme, is used as aprophylactic treatment in patients at risk of embolicstroke (Muir and Lees, 1997).

In this study, we used the Ab25-35-injected mouseas a model of AD to examine the mechanism ofaction by which cilostazol exerts its effects on learn-ing and memory impairment. Using both a sponta-neous alternation task and a step-down type passiveavoidance task, we compared the effects of cilostazolon learning and memory impairment induced byAb25-35 with those of aspirin, and then examined theeffect of cilostazol alone, using various dosingschedules, in more detail.

Methods

AnimalsSeven – 9-week-old male ddY mice (31–41 g, JapanSLC, Hamamatsu, Japan) were kept in a controlledenvironment for at least 5–7 days, with a 12-h light/12-h dark cycle (lights on; 7:45–19:45), and givenfood and tap water ad libitum. Experimental proto-cols concerning the use of laboratory animals wereapproved by the committee of Meijo University andwere performed in accordance with the guidelines ofthe Japanese Pharmacological Society (Folia Phar-macol. japon, 1992, 99: 35A) and the interministe-rial decree of 25 May 25 1987 (The Ministry ofEducation).

DrugsAmyloid b-protein fragment 25-35 (Ab25-35) was pur-chased from NeoMPS (NeoMPS, San Diego, CA,USA) and dissolved in distilled water at a concentra-tion of 1.8 mM. The Ab25-35 was incubated, or ‘aged’,by incubation in distilled water at 37°C for 4 days.Aged Ab25-35 (9 nmol 5 mL-1) or distilled water (5 mL)was injected i.c.v. as described previously (Mauriceet al., 1996) according to the methods of Haley and

BJP M Hiramatsu et al.

1900 British Journal of Pharmacology (2010) 161 1899–1912

McCormick (1957). The mice were lightly anaesthe-tized with ether, after which a 28-gauge stainless-steel needle was inserted into the lateral ventricle,and Ab25-35 was gradually injected over 30 s using amicroinfusion pump (KDS210, KD Scientific, Hollis-ton, MA, USA). The needle was left in place for 10 sfollowing the injection. The injection site was veri-fied in preliminary experiments by injecting bluedye solution. Neither insertion of the needle norinjection of the distilled water had a significantinfluence on behavioural responses or cognitivefunctions.

6-[4-(1-Cyclohexyl-1H-tetrazol-5-yl)butoxy]-3,4-dihydro-2(1H)-quinolinone (cilostazol) was ob-tained from Otsuka Pharmaceuticals, Inc. (Tokyo,Japan), and acetylsalicyclic acid (aspirin) wasobtained from Sigma (St. Louis, MO, USA). Drugswere suspended in 0.5% carboxymethyl cellulosesodium salt (CMC; Nacalai tesque, Kyoto, Japan)dissolved in distilled water. Cilostazol and aspirinwere administered p.o. 60 min before the behav-ioural tests. Sham control mice were administeredCMC, p.o. at a volume of 0.1 mL 10 g-1 body weight.

Experimental schedulesThe experimental schedules are shown in Figure 1.In Schedule #1, cilostazol and aspirin (30 and100 mg·kg-1, p.o.) were administered repeatedly for8 days following the injection of Ab25-35 (9 nmol permouse, i.c.v.). Spontaneous alternation performanceand the passive avoidance test were conducted onthe 5th and 7th days after Ab25-35 injection respec-tively. On these days, cilostazol and aspirin wereadministered 60 min before the behavioural tests. Inacute experiments, mice were only administered cil-ostazol or aspirin 60 min before these behaviouraltests on the 5th and 7th days. On the remainingdays, these mice were injected with 0.5% CMC solu-tion as reported in Schedule #1 (Schedule #2).

As the results of Schedule #1 revealed thatonly cilostazol ameliorated the learning and/ormemory impairments induced by Ab25-35, in the fol-lowing experiments only the effects of cilostazol,using various drug administration schedules, werestudied. The schedules were as follows: repeateddoses of cilostazol (100 mg·kg-1, p.o.) were adminis-tered from 8 days to 1 day before Ab25-35 injection(Schedule #3), from day 2 to 9 after Ab25-35 injection(Schedule #4), and from day 5 to 12 after Ab25-35

injection (Schedule #5). In these schedules, thespontaneous alternation performance was com-pleted on the 5th day, and the training trial of thepassive avoidance test was conducted on the 7th dayafter Ab25-35 injection (Schedules #1–3) or the firstinjection of cilostazol (Schedule #4). Each mousereceived eight injections of cilostazol in the repeated

experiments. On the 5th and 7th days, cilostazoland aspirin were administered 60 min prior to thebehavioural experiments. Sham control micereceived 0.5% CMC solution using these sameschedules.

Lipid peroxidation and cytokine levels weremeasured 2, 3, 5 and/or 7 days post-Ab peptide(25-35) (9 nmol per mouse, i.c.v.) injection for boththe acute and repeated cilostazol administrationschedules.

Spontaneous alternation behaviourImmediate working memory performance wasassessed by recording spontaneous alternationbehaviour during a single session in a Y-maze asdescribed previously (Itoh et al., 1993; Hiramatsuand Inoue, 1999), with minor modifications of theoriginal study performed in rats (Sarter et al., 1988).This paradigm includes exploratory, locomotor,motivational and systematic behaviours; the selec-tivity involving memory and especially learningprocesses is weak. However, this was used as a first-intent test because it is pharmacologically predictiveand is not constraining for animals. Furthermore,unlike the passive avoidance task, mice do notreceive electric shocks during the test periods. Ourprevious data indicated that there are certain corre-lations between the spontaneous alternation task inthe Y-maze and the passive avoidance task (Hira-matsu and Inoue, 2000a,b).

Each mouse, new to the maze, was placed at theend of one arm and allowed to move freely throughthe maze during an 8 min session. The series of armentries was recorded visually. Alternation wasdefined as the mouse entering different arms of themaze three times in succession as a result of con-secutive arm entering. The number of overlappingentrance sequences (e.g. ABC, BCA) was defined asthe number of alternations. The effect was calcu-lated as % alternation according to the followingformula:

% alternation number of alternations totalnumber of arm en

= (ttries 2− ) × 100

In this test, the animals must memorize the armentered immediately prior to the arm they are cur-rently occupying. Mice have a tendency to enter anewer environment, which in this case would be thearm not entered just before. Therefore, the % alter-nation is used as an index of short-term memory.The total number of arm entries is measured as anindex of locomotor activity, although it is not directevidence. Locomotor activity was also measured for10 min in control mice with Schedule #1 usingdigital counters with infrared sensors (MDC-W01,

BJPCilostazol prevents Ab-induced memory impairment

British Journal of Pharmacology (2010) 161 1899–1912 1901

Brain Science Idea, Co., Ltd., Osaka, Japan) in ablack Plexiglas box.

Step-down type passive avoidance taskA step-down type of passive avoidance task wasperformed as described previously (Hiramatsu andInoue, 2000a) with minor modification. The appa-ratus consisted of a transparent acrylic rectangularcage (30 ¥ 30 ¥ 40 cm high) comprised of a gridfloor with a wooden platform (4 ¥ 4 ¥ 4 cm) in the

centre set in a semi-soundproof wooden outer box(35 ¥ 35 ¥ 90 cm high). Illumination was providedby a 15 W lamp above the apparatus. An electriccurrent (1 Hz, 500 ms, 60 V, DC) was delivered tothe grid floor by an isolated stimulator (SEN-3201,Nihon Koden, Tokyo, Japan). The electrical resis-tance varied between 100 and 250 kW when micewere placed in the test cage. Therefore, each mousereceived an electric shock varying between 0.24and 0.60 mA.

Aβ peptide (25-35)(9 nmol, i.c.v.)





0 5 7 8 (days)

0 5 7 8 (days)

(days)

(days)

0

0

0 5 10 12 13

2 7 9 10

−1 5 7 8−8

(days)

Y-mazetest

Y-mazetest

Y-mazetest

Y-mazetest

Y-mazetest

Passiveavoidance test





Schedule #1(Figures 2 and 3)

Schedule #2(Figure 4)



Schedule #5

CMC (Vehicle), p.o.Cilostazol or aspirin (30, 100 mg kg−1, p.o.)

CMC (Vehicle), p.o.Cilostazol (30, 100 mg kg−1, p.o.)

CMC (Vehicle), p.o.Cilostazol (100 mg kg−1, p.o.)



Figure 1Experimental protocols used in this study. CMC, carboxymethyl cellulose sodium salt.



Each mouse was placed on the wooden platformfor the training test. When the mouse stepped downfrom the platform onto the grid floor, an electricshock was delivered for 15 s. A retention test wascarried out 24 h after the training session in amanner similar to the training, except that no elec-tric shock was delivered to the grid floor. Eachmouse was placed on the platform and the step-down latency was recorded. An upper cut-off time of300 s was set.

Responses to electric shockResponses to electric shock during the training ses-sions were recorded. The following scores weregiven based on the responses to electric shock: 3 =jumping, 2 = vocalization, 1 = flinching, 0 = noresponse. Shock sensitivity was shown as the totalscore, which was determined as the sum of eachscore following 15 s of recording.

Determination of lipid peroxidation levelMice were decapitated, and the hippocampus andfrontal cortex were removed. Brain samples werefrozen immediately by liquid N2 and stored at -80°Cin a deep freeze until assay.

Malondialdehyde (MDA) was measured with athiobarbituric acid-reactive substance assay kit(Cayman, Ann Arbor, MI, USA). Briefly, the isolatedbrain samples were homogenized in cooled RIPAbuffer (50 mM Tris-HCl buffer pH 7.4, 150 mMNaCl, 1% NP-40, 1% sodium deoxycholate, 0.1%SDS,1 mM PMSF, 1 mM Na3VO4, 1 mM NaF asprotease inhibitor cocktail) with an ultrasonichomogenizer (Microson Ultrasonic Cell Disruptor,Misonix, Farmingdale, NY, USA) for 15–20 s andcentrifuged at 4500¥ g for 10 min at 4°C. The homo-genate of brain samples was incubated with 8.1%sodium dodecylsulphate for 10 min followed by theaddition of 20% acetic acid (pH 3.5). The reactionmixture was incubated with 0.6% TBA in boilingwater for 2 h. After a 10 min cooling period in theice bath, the mixture was centrifuged at 1600¥ g for10 min at 4°C. The absorbance was determined by aplate reader (Wallac 1420 ARVOsx, Perkin Elmer,Waltham, MA, USA) at 550 nm. MDA content wasexpressed as mmol·mg-1 protein.

Determination of cytokine concentrationThe same tissue homogenates as those used for thelipid peroxidation assay were used for the cytokineassays (IL-1b, IL-2, IL-4, IL-5, IL-10, GM-CSF, IFN-gand TNF-a) using a multiplex bead-based immunoas-say kit (Mouse Cytokine 8-Plex A Assay kit, Bio-Rad,Hercules, CA, USA) according to the manufacturer’sprotocol. In brief, 50 mL of homogenate was platedinto a 96-well filter plate, coated with antibody-

coupled beads, and incubated for 1 h in a shadedroom using a platform shaker at ambient tempera-ture. The wells were then vacuum-filtered andwashed. Next, 50 mL of fluorescent solution wasadded, and the wells were incubated for 30 min. Thewells were again vacuum-filtered and washed, 125 mLof cytokine assay buffer was added, and the wellswere allowed to stand for 30 s before the intensity offluorescence was measured (Bio-Plex 200, Bio-Rad).

Protein assayThe protein content in diluted samples (¥50 and¥100) was measured based on a standard BSA usingthe Bio-Rad DC protein assay (Bio-Rad). Sampleswere read on a photometer (iMark MicroplateReader, Bio-Rad) set at 595 nm.

Data analysisThe behavioural data are expressed in terms ofmedian (vertical column) and interquartile rangesfrom the first to the third quartile (vertical line) forthe Y-maze test or box-plot (median and interquar-tile ranges) for the passive avoidance test. The datafor these memory tests were analysed using non-parametric type statistical methods because a Gaus-sian distribution was not always evident. Thus thesignificance of the differences was evaluated usingthe Mann–Whitney U-test for two group compari-sons or the Kruskal–Wallis test followed by the Bon-ferroni’s test for multiple comparisons.

The biochemical data are expressed as the means� SEM. Statistical differences among the experimen-tal groups were tested using two-way analysis ofvariance (ANOVA). Student’s t-test was used for com-parisons between two groups, and one-way ANOVA

followed by the Dunnett’s post hoc test was used formultiple comparisons. The criterion for significancewas set at P < 0.05 for all statistical evaluations. Allstatistical analyses were performed using the Prism 5Stat program (GraphPad Software, Inc., San Diego,CA, USA).

Results

Effects of repeated administration ofcilostazol and aspirin on Ab25-35-inducedimpairment of spontaneous alternationperformance in miceThe administration of Ab25-35 (9 nmol per mouse,i.c.v.) significantly decreased the % alternation inthe Y-maze when animals were tested 5 days later(Figure 2A,B). Repeated administration of cilostazol(30 and 100 mg·kg-1, p.o.) significantly and dose-dependently attenuated the impairment of sponta-neous alternation induced by Ab25-35 (Figure 2A).



Repeated administration of aspirin (30 and100 mg·kg-1, p.o.) did not alter the impairmentof spontaneous alternation induced by Ab25-35

(Figure 2B). Cilostazol and aspirin did not affect thetotal number of arm entries at these dosages (datanot shown). Repeated administration of cilostazol(100 mg·kg-1, p.o.) did not affect % alternation[control: 69.3 (64.5–75.4), cilostazol: 66.2 (58.1–70.2)] or locomotor activity (control: 302.0 (274.8–318.5), cilostazol: 255.0 (219.8–296.8)] in controlmice.

Effects of repeated administration ofcilostazol and aspirin on Ab25-35-inducedimpairment of learning and memory in miceusing the passive avoidance testThe step-down type passive avoidance test wasconducted to test the long-term memory of animalsadministered cilostazol. In the passive avoidancetest, Ab25-35 (9 nmol per mouse, i.c.v.) administrationsignificantly shortened the step-down latencywhen animals were tested 8 days after injection(Figure 3A,B). Repeated administration of cilostazol(30 and 100 mg·kg-1, p.o.) significantly improvedthe shortened step-down latency induced by Ab25-35

(Figure 3A). Compared with cilostazol, repeatedadministration of aspirin (30 and 100 mg·kg-1, p.o.)did not improve the learning and memory ofAb25-35-injected mice (Figure 3B). Cilostazol and

aspirin did not affect the response to electricshocks at these dosages (data not shown). Repeatedadministration of cilostazol (100 mg·kg-1, p.o.)did not prolong the step-down latency in controlmice [control: 82.3 (47.7–228.2), cilostazol: 119.0(30.5–252.5)].

Effects of acute administration of cilostazolon Ab25-35-induced behavioural impairmentAb25-35 (9 nmol per mouse, i.c.v.) significantlydecreased the % alternation and the step-downlatency 5 and 8 days after injections respectively.Acute administration of cilostazol (30 and100 mg·kg-1) did not improve these Ab25-35-inducedimpairments in learning and memory assessed inthe Y-maze and step-down type passive avoidancetests (Figure 4A,B).

Effects of pre-administration of cilostazol onAb25-35-induced behavioural impairmentCilostazol (100 mg·kg-1) was administered repeat-edly beginning 8 days before Ab25-35 injection to testwhether cilostazol can protect mice from Ab25-35

(9 nmol per mouse)-induced impairments inlearning and memory. Ab25-35 (9 nmol per mouse)significantly decreased the % alternation and thestep-down latency 5 and 8 days after injectionrespectively. However, pre-administration of cilosta-zol did not appear to protect mice from these


90

80

70

60

50

40

30

Y-mazetest


0 5 7 8 (days)Schedule #1

(Figures 2 and 3)


A B###

***

% A

ltern

atio

n

90

80

70

60

50

40

30

% A

ltern

atio

n

(13) (14) (12) (13) (13) (13) (13) (13)

Shamcontrol

Shamcontrol

Aβ peptide (25-35) Aβ peptide (25-35)30 100 30 100

Cilostazol (mg kg−1, p.o.) Aspirin (mg kg−1, p.o.)

Figure 2Effects of repeated administration of cilostazol (A) and aspirin (B) on Ab25-35-induced impairment of spontaneous alternation in the Y-maze test.Ab25-35 (9 nmol per mouse, i.c.v.) was injected 5 days before the Y-maze test. Mice were treated with cilostazol (30 and 100 mg·kg-1, p.o.) oraspirin (30 and 100 mg·kg-1, p.o.) once a day for 5 days. On the 5th day, these drugs were injected again 60 min before testing. Data are shownas median (vertical column) and first and third quartiles (vertical line). The number of mice used is shown in parentheses. Significant levels; *P <0.05, **P < 0.01 versus sham control (Mann–Whitney’s U-test). #P < 0.05, ##P < 0.01 versus Ab25-35 alone (Bonferroni’s test). CMC, carboxymethylcellulose sodium salt.




Y-mazetest


0 5 7 8 (days)Schedule #1

(Figures 2 and 3)


AB

Ste

p-do

wn

late

ncy

(s)

Ste

p-do

wn

late

ncy

(s)300

200

100

0

300

200

100

0

(13)

(13) (13) (13)

(13)

(13)

(13)

(14)

**##

#

Shamcontrol

Shamcontrol


Cilostazol (mg kg−1, p.o.) Aspirin (mg kg−1, p.o.)

*

Figure 3Effects of repeated administration of cilostazol (A) and aspirin (B) on Ab25-35-induced impairment of learning and memory in a passive avoidancetest. Ab25-35 (9 nmol per mouse, i.c.v.) was injected 7 days before the training trial. Mice were treated with cilostazol (30 and 100 mg·kg-1, p.o.)or aspirin (30 and 100 mg·kg-1, p.o.) once a day for 7 days. On the 7th day, these drugs were injected again 60 min before retention testing. Aretention trial was carried out 24 h after the training trial. Data are shown as median (horizontal bar) and first and third quartiles (vertical column).The number of mice used is shown in parentheses. Significant levels; *P < 0.05, **P < 0.01 versus sham control (Mann–Whitney’s U-test), #P < 0.05,##P < 0.01 versus Ab25-35 alone (Bonferroni’s test). CMC, carboxymethyl cellulose sodium salt.


Y-mazetest


0 5 7 8 (days)

CMC (Vehicle), p.o.Cilostazol (30, 100 mg kg−1, p.o.)


AB

90

80

70

60

50

40

30

% A

ltern

atio

n

****

(14) (14) (14) (14)

Shamcontrol

Shamcontrol


Cilostazol (mg kg−1, p.o.) Cilostazol (mg kg−1, p.o.)

Ste

p-do

wn

late

ncy

(s) 300

200

100

0

(13)

(13)

(13)

(13)

Figure 4Effects of acute administration of cilostazol on Ab25-35-induced impairment of spontaneous alternation (A) and learning and memory in a passiveavoidance test (B). Mice were injected with Ab25-35 (9 nmol per mouse, i.c.v.) 5 days before the training trial for the Y-maze test. Cilostazol (30 and100 mg·kg-1, p.o.) was injected 60 min before the Y-maze test and the training trial of the passive avoidance test. The retention trial was carriedout 24 h after the training trial. Data are shown as the median (vertical column) and first and third quartiles (vertical line) for the Y-maze andmedian (horizontal bar), and first and third quartiles (vertical column) for the passive avoidance tests. The number of mice in each experimentis shown in parentheses. Significant levels; **P < 0.01 versus sham control (Mann–Whitney’s U-test). CMC, carboxymethyl cellulose sodium salt.



deleterious effects of Ab25-35 on performance in theY-maze test and step-down type passive avoidancetest (Figure 5A,B).

Effects of repeated administration ofcilostazol on Ab25-35-induced behaviouralimpairmentsTo test whether cilostazol has protective and/orimproving effects on Ab25-35 (9 nmol per mouse)-induced impairments in behaviour, cilostazol(100 mg·kg-1) was administered 2 or 5 days after theAb25-35. Ab25-35 (9 nmol per mouse) significantlydecreased the % alternation 7 days after injectionand the step-down latency 10 days after injection(Figure 6A,B). Cilostazol significantly improvedAb25-35-induced impairment of spontaneous alterna-tion (Figure 6A) and tended to protect the micefrom the learning and memory impairment in thestep-down type passive avoidance test (Figure 6B,P = 0.063).

As described previously, an impairment of spon-taneous alternation was observed 5 days after Ab25-35

injection. Cilostazol administration was started 5days after Ab25-35 injection and continued for 6 daysin order to test whether cilostazol can have an effectafter the development of the impairment. Theresults indicated that cilostazol did not have an

ameliorating effect after repeated administrationunder this dosing schedule (Schedule #5) (data notshown).

Effect of cilostazol on MDA levels in thefrontal cortex and hippocampus ofAb25-35-injected miceThe effect of cilostazol on the levels of MDA in thefrontal cortex and hippocampus 2, 3, 5 and 7 daysafter Ab25-35 (9 nmol per mouse) injection wasexamined to determine whether lipid peroxidationis involved in its ameliorating effect in these mice.A significant increase in the levels of MDA wasobserved in both brain areas of Ab25-35-injected micecompared with the control group (F(1,41) = 38.75,P < 0.01 and F(1,44) = 16.55, P < 0.01, respectively,Figure 7). The level of MDA increase was dependentupon the numbers of days that had passed sincethe Ab25-35. The levels of MDA were significantlyhigher at days 2, 3 and 5 post-Ab25-35 injection inthe frontal cortex and at 3 and 5 days after Ab25-35

injection in the hippocampus compared withthe control group. However, at 7 days after Ab25-35

injection, the increase in MDA levels in bothbrain regions returned to the control levels.Repeated treatment with cilostazol (100 mg·kg-1)prevented the increase in MDA levels in the frontal



Y-mazetest


0−1−8 5 7 8(days)


A B**

**90

80

70

60

50

40

30

% A

ltern

atio

n

Shamcontrol

Shamcontrol

Aβ peptide (25-35) Aβ peptide (25-35)100 100


(19) (19)

(19)

(16)

(16)

(18)

Ste

p-do

wn

late

ncy

(s) 300

200

100

0

Figure 5Effects of repeated pre-administration of cilostazol on Ab25-35-induced impairment of spontaneous alternation (A) and learning and memory in apassive avoidance test (B). Mice were treated with cilostazol (100 mg·kg-1, p.o.) repeatedly for 8 days beginning 8 days before Ab25-35 (9 nmol permouse, i.c.v.) injection. The Y-maze test was carried out 5 days after (A) Ab25-35 injection, and the passive avoidance test was carried out 7–8 daysafter Ab25-35 injection (B). The retention trial was carried out 24 h after the training trial. Data are shown as the median (vertical column) and firstand third quartiles (vertical line) for the Y-maze, and median (horizontal bar) and first and third quartiles (vertical column) for the passive avoidancetests. The number of mice used is shown in parentheses. Significant levels; **P < 0.01 versus sham control (Mann–Whitney’s U-test). CMC,carboxymethyl cellulose sodium salt.



cortex and hippocampus of Ab25-35-treated mice(Figure 7A,B). In contrast, acute treatment withcilostazol (100 mg·kg-1) had no effect on thechanges in MDA levels induced by Ab25-35 injection(Figure 7A,B). In control mice, repeated administra-tion of cilostazol (100 mg·kg-1, p.o.) for 5 daysslightly increased and decreased MDA levels inthe frontal cortex (control: 1.23 � 0.05, cilostazol:1.41 � 0.04 nmol·mg-1 protein) and hippocam-pus (control: 1.03 � 0.04, cilostazol: 0.91 �0.03 nmol·mg-1 protein) respectively.

Effects of cilostazol on cytokine levels in thehippocampus of Ab25-35-injected miceTo determine whether pro-inflammatory cytokinesare involved in the lipid peroxidation and/or theameliorating effect of cilostazol in Ab25-35-treatedmice, we examined the effect of cilostazol on thelevels of cytokines [IL-1b, IL-2, IL-4 and granulocytemacrophage colony-stimulating factor (GM-CSF)] inthe hippocampus 2, 3 and 5 days after administra-tion of Ab25-35 (9 nmol per mouse). A significantincrease in the levels of IL-1b was observed in thehippocampus of Ab25-35-injected mice (F(1,38) =9.447, P < 0.01, Figure 8). In the group of 5 dayspost-injection, the levels of IL-1b significantlyincreased compared with the control group(P < 0.05). Repeated treatment with cilostazol

(100 mg·kg-1) did not prevent this increase in IL-1blevels in the hippocampus of Ab25-35-treated mice(Figure 8). Further, cilostazol had no effect on thelevels of other cytokines, including IL-2, IL-4 andGM-CSF. Acute treatment with cilostazol also had noeffect on IL-1b levels after Ab25-35 injection (Figure 8).

Discussion

In the present study, Ab25-35 induced memoryimpairment in both the Y-maze and the step-downtype passive avoidance tests. These results are con-sistent with previous reports that Ab25-35 inducescognitive impairment in mice (Maurice et al., 1996;Hiramatsu et al., 2000; Alkam et al., 2007; 2008).Using this model, the effects of cilostazol onmemory impairment were examined. Cilostazol pre-vented Ab25-35-induced short-term and long-termmemory impairment in the Y-maze and the step-down type passive avoidance tests respectively. Cil-ostazol had no effect on locomotor activity, assessedas the number of arm entries during the Y-maze testand ambulation in the locomotor test, or on shocksensitivity, assessed as the total score in the passiveavoidance test. These results suggest that cilostazolattenuates cognitive impairments in the Ab25-35-injected mice without affecting motor function,



Y-mazetest


0 2 7 9 10 (days)


A B##** *90

80

70

60

50

40

30

% A

ltern

atio

n

(14) (14) (14)

Shamcontrol

Shamcontrol

Aβ peptide (25-35) Aβ peptide (25-35)100 100


Ste

p-do

wn

late

ncy

(s) 300

200

100

0

(16)

(15)(15)

Figure 6Effects of repeated administration of cilostazol on Ab25-35-induced impairment of spontaneous alternation (A) and learning and memory in a passiveavoidance test (B). Mice were injected with Ab25-35 (9 nmol per mouse, i.c.v.) 7 days before the Y-maze test or 9 days before the passive avoidancetest. Mice were treated with cilostazol (100 mg·kg-1, p.o.) repeatedly for 5 days beginning 2 days after Ab25-35 (9 nmol per mouse, i.c.v.) injection.Cilostazol was administered 60 min before testing on the Y-maze test day. On the next day, cilostazol was injected. On the day of the training trialfor the passive avoidance test, cilostazol was administered 60 min before testing. A retention trial was carried out 24 h after the training trial. Dataare shown as the median (vertical column) and first and third quartiles (vertical line) for the Y-maze, and median (horizontal bar) and first and thirdquartiles (vertical column) for the passive avoidance tests. The number of mice in each experiment is shown in parentheses. Significant levels;*P < 0.05, **P < 0.01 versus sham control, #P < 0.05 versus Ab25-35 alone (Mann–Whitney’s U-test). CMC, carboxymethyl cellulose sodium salt.



motivation, exploratory activity and/or shocksensitivity. Furthermore, cilostazol prevented theaccumulation of lipid peroxide (MDA level) in thefrontal cortex and hippocampus in the early periodafter Ab25-35 treatment, as MDA levels in both regionsreturned to control levels by 7 days after Ab25-35

treatment. To our knowledge, this is the first reportstating that cilostazol protects against Ab25-35-induced neurotoxicity, maybe involving oxidativestress in the brain. This finding is interestingbecause cilostazol has been shown to have beneficial

effects in clinical situations, improving the memoryfunction of Alzheimer patients taking donepezil(Arai and Takahashi, 2009). In a recent study, it wasshown that donepezil (5 mg·kg-1, p.o.) taken for 13days following okadaic acid injection improvedmemory performance and also significantly restoredMDA and glutathione levels in rats (Kamat et al.,2010). Lee et al. (2007) reported that concurrentadministration of cilostazol (30 mg·kg-1, p.o.) withdonepezil (0.3 mg·kg-1, i.p.) effectively attenuatescognitive dysfunction and increases neuroprotec-tion after chronic cerebral hypoperfusion in rats.Further, Okajima and his colleagues have recentlyreported that donepezil and cilostazol improve cog-nitive function in mice by increasing the produc-tion of insulin-like growth factor-1 (IGF-1)(Narimatsu et al., 2009; Zhao et al., 2010). Takentogether, these findings indicate that cilostazol canhave a beneficial effect not only on vascular demen-tia, but also in AD.

It has been reported that oxidative stress plays acritical role in the development of AD and mildcognitive impairment (Markesbery, 1997; Mecocci,2004). Lipid peroxidation is one of the major out-comes of free radical-mediated injury; it directlydamages membranes and generates a number ofsecondary products including aldehydes, such asMDA (Slater, 1984). Further, analysis of brains ofAD patients has demonstrated that they have anincrease in lipid peroxidation products in the hip-pocampus and amygdala compared with age-matched controls (Markesbery and Lovell, 1998).Amyloid precursor protein transgenic mice, agenetic mouse model of AD, display a systemicincrease in lipid peroxidation (Praticò et al., 2001).In addition, lipid peroxidation up-regulates theb-site of amyloid precursor protein (APP) cleavingenzyme 1 (BACE1) expression in vivo (Chen et al.,2008), suggesting that prevention of lipid peroxida-tion is a very important early event for amyloido-genesis in AD. Because MDA is the most abundantindividual aldehyde resulting from lipid peroxida-tion, it is a useful oxidative marker for AD and mildcognitive impairment (Greilberger et al., 2008). Toconfirm the effect of cilostazol on Ab25-35-inducedoxidative stress, we measured the levels of MDA (amarker of lipid peroxidation) in the frontal cortexand hippocampus. Ab25-35 significantly increased thelevels of MDA in the frontal cortex and hippocam-pus 2, 3 and/or 5 days after injection. Interestingly,there was no significant increase in MDA levels inthe Ab25-35-treated group on day 7. This suggests thatit is very important to prevent oxidative stress in theearly period after Ab25-35 treatment. After Ab25-35

treatment, oxidative stress increases during days 1–5(in the case of hippocampus, days 3–5) and this may

Figure 7Effects of cilostazol on Ab25-35-induced increase in lipid peroxidationin the mouse frontal cortex (A) and hippocampus (B). Ab25-35 (9 nmolper mouse) was injected (i.c.v.) and the mice were killed 2, 3, 5and/or 7 days later. Mice were treated with cilostazol (100 mg·kg-1,p.o.) 60 min before being killed (acute) or repeatedly (3, 4, 6 or 8times). On the 1st day, cilostazol was administered 60 min beforeAb25-35. Lipid peroxidation levels were assessed by the thiobarbituricacid method. Each lipid peroxidation level was normalized per unitprotein. Values represent means � SEM (n = 6–9). Significant levels;*P < 0.05, **P < 0.01 versus sham control (Student’s t-test), ##P <0.01 versus Ab25-35 alone (Dunnett’s test). MDA, malondialdehyde.



affect subsequent behavioural events. Thereafter,oxidative stress may decrease defensive systems, atthe next step contributing to behavioural impair-ment or, simply, a single injection of Ab25-35 does nothave a long-lasting effect on lipid peroxidation. Thiscould explain why cilostazol did not have a signifi-cant ameliorating effect when it was administered 5days after the Ab25-35 injection. Repeated treatmentwith cilostazol completely prevented the accumula-tion of MDA induced by Ab25-35, while acute treat-ment did not show any effect. The level of MDA inthe hippocampus 2 days after Ab25-35 injection wasnot significantly different from control levels.Behaviourally, only the schedule in which cilostazolwas injected repeatedly, starting from 1 or 2 days,but not 5 days, after Ab25-35 administration, signifi-cantly prevented the memory impairments inducedby this peptide. These results suggest that the pro-tective effect of cilostazol on Ab25-35-inducedmemory impairment is related to the accumulationof oxidative stress in the early period after Ab25-35

injection, in part in the frontal cortex and hippoc-ampus. This indicates that not only oxidative stress,but also other mechanisms, may be involved in theameliorative effect of cilostazol. In fact, as describedpreviously, it has been reported that cilostazolimproved cognitive function by increasing the

production of IGF-1 in the hippocampus (Zhaoet al., 2010). Therefore, we need to check whetherIGF-1 production in the hippocampus and/orfrontal cortex is involved in this ameliorating effectof cilostazol.

It has been reported that Ab25-35 increases themRNA levels of APP and IL-1b. Consistent withthese changes in mRNA levels, immunoblottinganalysis revealed that Ab25-35 also increases APP,BACE and IL-1b in the rat hippocampus (Lin et al.,2009). Previous research in our laboratory hasshown that lipopolysaccharide (LPS) administrationproduces learning and/or memory deficits in avariety of paradigms along with an increase in IL-1b.ROS are also generated following LPS administra-tion, indicating that there are some commonmechanisms between these models. ROS release byactivated microglia, which are present near amyloidfibres, has been reported in the brains of patientswith AD (Tarkowski et al., 2001). The mechanism bywhich ROS are generated by Ab1-40 involves the pro-duction of cytokines by activated microglia thatthen act on nuclear factor kB, inhibitor of NF-kB orother transcription factors leading to the inductionof iNOS (Medeiros et al., 2007). The production ofcytokines by these microglia, as well as lipid peroxi-dation, may be associated with the onset and pro-

Figure 8Effects of cilostazol on the levels of pro-inflammatory cytokines in mouse hippocampus after Ab25-35 injection. Ab25-35 (9 nmol per mouse) wasinjected (i.c.v.) and the mice were killed 2, 3 and/or 5 days later. Mice were treated with cilostazol (100 mg·kg-1, p.o.) 60 min prior to death(acute) or repeatedly (3, 4 or 6 times). On the 1st day, cilostazol was administered 60 min before Ab25-35. Pro-inflammatory cytokine levels wereassessed by the suspension array method. Each cytokine level was normalized per unit protein. Values represent means � SEM (n = 4–6).Significant levels; *P < 0.05 versus corresponding sham control (Student’s t-test). GM-CSF, granulocyte macrophage colony-stimulating factor.



gression of the dementia associated with AD.However, in the present study, repeated administra-tion of cilostazol not only failed to antagonize theincrease in the expression of IL-1b, but there wasalso no significant difference in the expression ofIL-2, IL-4 or GM-CSF after Ab25-35 injection. There-fore, there are some differences between theseanimal models, and cilostazol may only provideprotection against oxidative stress.

Although the mechanism by which cilostazolregulates Ab25-35-induced oxidative stress remains tobe determined in detail, there are several possibleexplanations. DNA fragmentation in the corticaltissue of rats with focal ischaemia has been shownto be significantly suppressed by cilostazol, but notby aspirin (Lee et al., 2005). This suggests that cil-ostazol, but not aspirin, has an anti-apoptotic effect.Lee et al. (2004) elucidated the signalling pathwayby which cilostazol suppresses increased phos-phatase and tensin homolog deleted from chromo-some 10 (PTEN) phosphorylation, which is evokedfollowing 2 h of middle cerebral artery occlusion/24 h reperfusion, as being in association withreduced apoptotic cell death in rats. Cilostazol has abroad spectrum of activities, including antagonismof PDE III. The mode of action of cilostazol is differ-ent from that of aspirin in that cilostazol inhibitsnot only secondary platelet aggregation but alsoprimary platelet aggregation induced by aggregatingagents such as ADP. Cilostazol is potent at prevent-ing death due to pulmonary thrombosis by plateletaggregates in mice in vivo (Kimura et al., 1985). Cil-ostazol also increases intracellular cyclic AMP levelsby inhibiting PDE III, scavenging hydroxyl radicals,and reducing intracellular ROS and TNF-a produc-tion induced by LPS (Kim et al., 2002) and chroniccerebral hypoperfusion (Lee et al., 2006). Further-more, it has been reported that cilostazol exerts aprotective effect in the brain through the cAMP-responsive element binding protein (CREB) phos-phorylation pathway leading to the up-regulation ofBcl-2 and COX-2 expression (Kim et al., 2006) afterchronic cerebral hypoperfusion. These data suggestthat cilostazol is potentially useful for the treatmentof cognitive impairment in post-stroke patients(Watanabe et al., 2006). Recently, Miyamoto et al.(2009) reported that phosphorylated CREB withinneuroblasts is markedly decreased in the subven-tricular zone and olfactory bulb of vehicle-treatedrats under conditions of chronic cerebral hypoper-fusion. However, treatment with cilostazol resultedin recovery of the expression of this CREB through-out the hypoperfusion period, leading to enhancedneurogenesis. Further, cilostazol improved cognitivefunction by increasing the production of IGF-1 inthe hippocampus (Zhao et al., 2010). In line with

these findings, it is possible that cilostazol not onlyprevents oxidative damage directly by scavengingfree radicals but may also work via other mecha-nisms such as CREB signalling, which may be a keymediator of neurogenesis. While we have used acutetreatment of mice with the Ab25-35 as a model for ADin the present study, it may not be the best modelfor AD; therefore, the effects of cilostazol need to bereplicated in more physiological models of AD, andthe involvement of additional molecular correlatesneeds to be clarified, such as CREB signalling, DNAfragmentation, as an antiapoptotic effect and IGF-1production.

In conclusion, the results from the present studyconfirm, for the first time, that cilostazol amelio-rates the memory deficits induced by Ab25-35 in mice.The effect of cilostazol may be attributed to theprevention of oxidative damage in the hippocam-pus as measured in terms of the amount of peroxi-dized lipid. Cilostazol is well tolerated as atherapeutic agent, and is itself, or in combinationwith other drugs such as donepezil, a potential can-didate for the treatment of cognitive deficits in AD.

Acknowledgements

This study was supported by Otsuka PharmaceuticalCo., Ltd. and in part by the Academic FrontierProject for Private Universities (2007-11) from theMinistry of Education, Culture, Sports, Science andTechnology of Japan.

Conflict of interest

None.

References

Alkam T, Nitta A, Mizoguchi H, Itoh A, Nabeshima T(2007). A natural scavenger of peroxynitrites, rosmarinicacid, protects against impairment of memory inducedby Abeta(25-35). Behav Brain Res 180: 139–145.

Alkam T, Nitta A, Mizoguchi H, Itoh A, Murai R,Nagai T et al. (2008). The extensive nitration ofneurofilament light chain in the hippocampus isassociated with the cognitive impairment induced byamyloid beta in mice. J Pharmacol Exp Ther 327:137–147.

Arai H, Takahashi T (2009). A combination therapy ofdonepezil and cilostazol for patients with moderateAlzheimer disease: pilot follow-up study. Am J GeriatrPsychiatry 17: 353–354.



Chen L, Na R, Gu M, Richardson A, Ran Q (2008). Lipidperoxidation up-regulates BACE1 expression in vivo: apossible early event of amyloidogenesis in Alzheimer’sdisease. J Neurochem 107: 197–207.

Choi JM, Shin HK, Kim KY, Lee JH, Hong KW (2002).Neuroprotective effect of cilostazol against focal cerebralischemia via antiapoptotic action in rats. J PharmacolExp Ther 300: 787–793.

Fang F, Liu GT (2006). Protective effects of compoundFLZ on beta-amyloid peptide-(25-35)-induced mousehippocampal injury and learning and memoryimpairment. Acta Pharmacol Sin 27: 651–658.

Greilberger J, Koidl C, Greilberger M, Lamprecht M,Schroecksnadel K, Leblhuber F et al. (2008).Malondialdehyde, carbonyl proteins andalbumin-disulphide as useful oxidative markers in mildcognitive impairment and Alzheimer’s disease. FreeRadic Res 42: 633–638.

Haley TJ, McCormick WG (1957). Pharmacologicaleffects produced by intracerebral injection of drugs inthe conscious mouse. Br J Pharmacol Chemother 12:12–15.

Hiramatsu M, Inoue K (1999). Nociceptin/orphanin FQand nocistatin on learning and memory impairmentinduced by scopolamine in mice. Br J Pharmacol 127:655–660.

Hiramatsu M, Inoue K (2000a). Des-tyrosine(1)dynorphin A-(2-13) improves carbon monoxide-inducedimpairment of learning and memory in mice. Brain Res859: 303–310.

Hiramatsu M, Inoue K (2000b). Improvement by lowdoses of nociceptin on scopolamine-inducedimpairment of learning and/or memory. Eur JPharmacol 395: 149–156.

Hiramatsu M, Inoue K, Kameyama T (2000). DynorphinA-(1-13) and (2-13) improve beta-amyloidpeptide-induced amnesia in mice. Neuroreport 11:431–435.

Itoh J, Ukai M, Kameyama T (1993). Dynorphin A-(1-13)markedly improves scopolamine-induced impairment ofspontaneous alternation performance in mice. Eur JPharmacol 236: 341–345.

Kamat PK, Tota S, Saxena G, Shukla R, Nath C (2010).Okadaic acid (ICV) induced memory impairment in rats:a suitable experimental model to test anti-dementiaactivity. Brain Res 1309: 66–74.

Kim KY, Shin HK, Choi JM, Hong KW (2002). Inhibitionof lipopolysaccharide-induced apoptosis by cilostazol inhuman umbilical vein endothelial cells. J PharmacolExp Ther 300: 709–715.

Kim MJ, Lee JH, Park SY, Hong KW, Kim CD, Kim KYet al. (2006). Protection from apoptotic cell death bycilostazol, phosphodiesterase type III inhibitor, viacAMP-dependent protein kinase activation. PharmacolRes 54: 261–267.

Kimura Y, Tani T, Kanbe T, Watanabe K (1985). Effect ofcilostazol on platelet aggregation and experimentalthrombosis. Arzneimittelforschung 35: 1144–1149.

Kwon SU, Cho YJ, Koo JS, Bae HJ, Lee YS, Hong KS et al.(2005). Cilostazol prevents the progression of thesymptomatic intracranial arterial stenosis: themulticenter double-blind placebo-controlled trial ofcilostazol in symptomatic intracranial arterial stenosis.Stroke 36: 782–786.

Lee JH, Kim KY, Lee YK, Park SY, Kim CD, Lee WS et al.(2004). Cilostazol prevents focal cerebral ischemic injuryby enhancing casein kinase 2 phosphorylation andsuppression of phosphatase and tensin homolog deletedfrom chromosome 10 phosphorylation in rats. JPharmacol Exp Ther 308: 896–903.

Lee JH, Park SY, Lee WS, Hong KW (2005). Lack ofantiapoptotic effects of antiplatelet drug, aspirin andclopidogrel, and antioxidant, MCI-186, against focalischemic brain damage in rats. Neurol Res 27: 483–492.

Lee JH, Park SY, Shin YW, Hong KW, Kim CD, Sung SMet al. (2006). Neuroprotection by cilostazol, aphosphodiesterase type 3 inhibitor, against apoptoticwhite matter changes in rat after chronic cerebralhypoperfusion. Brain Res 1082: 182–191.

Lee JH, Park SY, Shin YW, Kim CD, Lee WS, Hong KW(2007). Concurrent administration of cilostazol withdonepezil effectively improves cognitive dysfunctionwith increased neuroprotection after chronic cerebralhypoperfusion in rats. Brain Res 1185: 246–255.

Lin HB, Yang XM, Li TJ, Cheng YF, Zhang HT, Xu JP(2009). Memory deficits and neurochemical changesinduced by C-reactive protein in rats: implication inAlzheimer’s disease. Psychopharmacology (Berl) 204:705–714.

Lu P, Mamiya T, Lu LL, Mouri A, Zou L, Nagai T et al.(2009). Silibinin prevents amyloid beta peptide-inducedmemory impairment and oxidative stress in mice. Br JPharmacol 157: 1270–1277.

Markesbery WR (1997). Oxidative stress hypothesis inAlzheimer’s disease. Free Radic Biol Med 23: 134–147.

Markesbery WR, Lovell MA (1998).Four-hydroxynonenal, a product of lipid peroxidation,is increased in the brain in Alzheimer’s disease.Neurobiol Aging 19: 33–36.

Maurice T, Lockhart BP, Privat A (1996). Amnesiainduced in mice by centrally administered beta-amyloidpeptides involves cholinergic dysfunction. Brain Res706: 181–193.

Mecocci P (2004). Oxidative stress in mild cognitiveimpairment and Alzheimer disease: a continuum. JAlzheimers Dis 6: 159–163.

Medeiros R, Prediger RD, Passos GF, Pandolfo P,Duarte FS, Franco JL et al. (2007). Connecting TNF-alphasignaling pathways to iNOS expression in a mousemodel of Alzheimer’s disease: relevance for the



behavioral and synaptic deficits induced by amyloidbeta protein. J Neurosci 27: 5394–5404.

Meunier J, Ieni J, Maurice T (2006). The anti-amnesicand neuroprotective effects of donepezil against amyloidbeta25-35 peptide-induced toxicity in mice involve aninteraction with the sigma1 receptor. Br J Pharmacol149: 998–1012.

Miyamoto N, Tanaka R, Zhang N, Shimura H,Onodera M, Mochizuki H et al. (2009). Crucial role forSer133-phosphorylated form of cyclic AMP-responsiveelement binding protein signaling in the differentiationand survival of neural progenitors under chroniccerebral hypoperfusion. Neuroscience 162: 525–536.

Montine TJ, Markesbery WR, Zackert W, Sanchez SC,Roberts LJ 2nd, Morrow JD (1999). The magnitude ofbrain lipid peroxidation correlates with the extent ofdegeneration but not with density of neuritic plaques orneurofibrillary tangles or with APOE genotype inAlzheimer’s disease patients. Am J Pathol 155: 863–868.

Muir K, Lees KR (1997). Thrombolytic therapy for acuteischaemic stroke. Lancet 350: 1476–1477.

Narimatsu N, Harada N, Kurihara H, Nakagata N,Sobue K, Okajima K (2009). Donepezil improvescognitive function in mice by increasing the productionof insulin-like growth factor I in the hippocampus. JPharmacol Exp Ther 330: 2–12.

Palmer AM (2002). Pharmacotherapy for Alzheimer’sdisease: progress and prospects. Trends Pharmacol Sci23: 426–433.

Pappolla MA, Chyan YJ, Omar RA, Hsiao K, Perry G,Smith MA et al. (1998). Evidence of oxidative stress andin vivo neurotoxicity of beta-amyloid in a transgenicmouse model of Alzheimer’s disease: a chronic oxidativeparadigm for testing antioxidant therapies in vivo. Am JPathol 152: 871–877.

Praticò D, Uryu K, Leight S, Trojanoswki JQ, Lee VM(2001). Increased lipid peroxidation precedes amyloidplaque formation in an animal model of Alzheimeramyloidosis. J Neurosci 21: 4183–4187.

Rowan MJ, Klyubin I, Wang Q, Hu NW, Anwyl R (2007).Synaptic memory mechanisms: Alzheimer’s diseaseamyloid beta-peptide-induced dysfunction. Biochem SocTrans 35 (Pt 5): 1219–1223.

Sarter M, Bodewitz G, Stephens DN (1988). Attenuationof scopolamine-induced impairment of spontaneousalteration behaviour by antagonist but not inverseagonist and agonist beta-carbolines.Psychopharmacology (Berl) 94: 491–495.

Slater TF (1984). Free-radical mechanisms in tissueinjury. Biochem J 222: 1–15.

Sugaya K, Uz T, Kumar V, Manev H (2000). Newanti-inflammatory treatment strategy in Alzheimer’sdisease. Jpn J Pharmacol 82: 85–94.

Szekely CA, Green RC, Breitner JC, Østbye T, Beiser AS,Corrada MM et al. (2008). No advantage of A beta42-lowering NSAIDs for prevention of Alzheimerdementia in six pooled cohort studies. Neurology 70:2291–2298.

Tarkowski E, Wallin A, Regland B, Blennow K,Tarkowski A (2001). Local and systemic GM-CSF increasein Alzheimer’s disease and vascular dementia. ActaNeurol Scand 103: 166–174.

Tran MH, Yamada K, Olariu A, Mizuno M, Ren XH,Nabeshima T (2001). Amyloid beta-peptide inducesnitric oxide production in rat hippocampus: associationwith cholinergic dysfunction and amelioration byinducible nitric oxide synthase inhibitors. FASEB J 15:1407–1409.

Trubetskaya VV, Stepanichev MY, Onufriev MV,Lazareva NA, Markevich VA, Gulyaeva NV (2003).Administration of aggregated beta-amyloid peptide(25-35) induces changes in long-term potentiation inthe hippocampus in vivo. Neurosci Behav Physiol 33:95–98.

Watanabe T, Zhang N, Liu M, Tanaka R, Mizuno Y,Urabe T (2006). Cilostazol protects against brain whitematter damage and cognitive impairment in a rat modelof chronic cerebral hypoperfusion. Stroke 37:1539–1545.

Zhao J, Harada N, Kurihara H, Nakagata N, Okajima K(2010). Cilostazol improves cognitive function in miceby increasing the production of insulin-like growthfactor I in the hippocampus. Neuropharmacology 58:774–783.



Cilostazol prevents amyloid β peptide25-35-induced memory impairment and oxidative stress in mice

Documents

Transcript of Cilostazol prevents amyloid β peptide25-35-induced memory impairment and oxidative stress in mice